Open Access

ARTICLE

The Machine Learning Ensemble for Analyzing Internet of Things Networks: Botnet Detection and Device Identification

Seung-Ju Han, Seong-Su Yoon, Ieck-Chae Euom^*

System Security Research Center, Chonnam National University, Gwangju, 61186, Republic of Korea

* Corresponding Author: Ieck-Chae Euom. Email:

(This article belongs to the Special Issue: Advanced Security for Future Mobile Internet: A Key Challenge for the Digital Transformation)

Computer Modeling in Engineering & Sciences 2024, 141(2), 1495-1518. https://doi.org/10.32604/cmes.2024.053457

Received 30 April 2024; Accepted 24 July 2024; Issue published 27 September 2024

Abstract

The rapid proliferation of Internet of Things (IoT) technology has facilitated automation across various sectors. Nevertheless, this advancement has also resulted in a notable surge in cyberattacks, notably botnets. As a result, research on network analysis has become vital. Machine learning-based techniques for network analysis provide a more extensive and adaptable approach in comparison to traditional rule-based methods. In this paper, we propose a framework for analyzing communications between IoT devices using supervised learning and ensemble techniques and present experimental results that validate the efficacy of the proposed framework. The results indicate that using the proposed ensemble techniques improves accuracy by up to 1.7% compared to single-algorithm approaches. These results also suggest that the proposed framework can flexibly adapt to general IoT network analysis scenarios. Unlike existing frameworks, which only exhibit high performance in specific situations, the proposed framework can serve as a fundamental approach for addressing a wide range of issues.

---

Keywords

---